Semantic-aware Resource Allocation for Wireless Image Transmission

作者：Xue Han, Biqian Feng, Yuxuan Shi, Yongpeng Wu, Wenjun Zhang · 年份：2024 · DOI：10.1109/iccc62479.2024.10681891 · 被引用次数：6 · 研究领域：Advanced Data Compression Techniques、Video Coding and Compression Technologies

As a promising technology for future communication systems, semantic communication is expected to guarantee transmission reliability and communication efficiency. In this paper, we investigate a semantic-aware resource allocation problem within a typical multi-cell multi-user multiple-input multiple-output (MIMO) semantic communication system, aiming to efficiently transmit images. Specifically, we formulate the joint optimization of transmission latency and task performance resource allocation problem in terms of transceiver beamformer design, the number of transmitted semantic symbols, and channel assignment. To solve this problem, we adopt the alternative optimization method and decouple it into three subproblems with other variables fixed. Then we take successive convex approximation, exhaustive search, and Hungarian algorithm to solve these subproblems, respectively. Simulation results demonstrate the effectiveness of the proposed resource allocation method, as well as the superiority of semantic communication in terms of transmission latency and inference performance in downstream tasks.